Power analysis systems in residences have not been successful in the marketplace because the systems require sensors on resource consuming devices, like refrigerators, pool pumps and the like. The costs of the sensors and their installation can be so high that homeowners are unwilling to take the risk of purchasing a system that may not save them more than the cost of the system.
As smart power meters are installed between the power grid and homes, this represents an opportunity to install a solution in the power meter that can be used to implement power analysis functions. For example, a meter could be developed that would recognize a power signature of each device in the home. However, this would require a meter that reported the power thousands of times per second, to provide the resolution of the power use that could be used in such a system. Currently available meters do not report this frequently. Although they could be made to increase their reporting rate, this too would increase the cost of the system. Utilities are frequently the party that bears the cost of a retrofit of a meter. Many utilities would not want to pay the extra costs for such a meter. Furthermore, the network capacity for many meters may be limited, and that much data could be problematic.
What is needed is a system and method that can analyze power usage without requiring power sensors on every device, can use the capabilities of existing smart meters, and will not overload the network bandwidth of the smart meters.